Ellison C. Pierce, Jr., MD, is Chairman of the Department of Anesthesia, New England Deaconess Hospital, Boston, and Associates Professor at Harvard Medical School
The first reported death from Anesthesia was that of a 15-yearold British girl given chloroform for toenail removal in 1848, only 15 months after William Morton first demonstrated the use of ether in Boston. Dr. John Snow, the first true anesthesiologist, maintained that the girl’s death was caused by the direct action of chloroform on her heart.1
Controversy concerning the relative safety of chloroform versus ether lasted some 60 years until it was shown that the combination of light not deep chloroform and its associated increase in circulating epinephrine could produce ventricular fibrillation. Nevertheless, chloroform continued to be used until the middle of this century. Dr. Snow also described the need to measure pulse and respiration rates to determine both depth of anesthesia and onset of circulatory failure, but even today the extent to which a patient undergoing anesthesia should be monitored is still debated.
In the United States, anesthesia is administered some 20 million times each year. The incidence of directly related mortality is believed to lie in the range of 1 per 10,000 to 1 per 4,000 administrations-some 2,0005,000 deaths each year.2 Most writers suggest that most of these deaths are preventable and that they frequently involve human error. Keenan and Boyan, in a recent review, found the incidence of cardiac arrest due to anesthesia to be 1.7 per 10,000 administrations in their hospital, with a resultant mortality rate of 0.9 per I 0,000. In their study, they were able to identify a specific anesthetic error in 75% of the occurrences.
In other reviews, ECRI (Emergency Care Research Institute, a nonprofit biomedical engineering research association) and Pierce have outlined multifaceted approaches that must be taken if the incidence of avoidable anesthetic mishaps is to be reduced.4,1 Duberman and Bendixen have emphasized that today the public expects near-zero rates of morbidity and mortality in generally healthy patients undergoing anesthesia.6
Seriousness of the Problem
Although anesthetic medical liability claims make up only about 3% or 4% of the total in medicine, the indemnity paid exceeds 10%, a fact obviously related to the severity of the injuries. In reviews of closed and open anesthesia-related claim abstracts, certain types of problems clearly recur. 7,8 Variations in classification make comparisons difficult, but common complications include inadequate ventilation, difficult endotracheal intubation, esophageal intubation, inadvertent extubation, ventilator disconnects, relative or absolute overdosages, and bronchospasm.
In addition to the unknown number of anesthetic deaths each year, there are other anesthesia related mishaps -perhaps as many as those leading to death-that result in severe morbidity such as permanent brain damage.
Eliminating anesthesia mishaps is as important as changing tort law to control the cost of medical liability insurance. Proper monitoring is a major nonlegal approach to this problem because backing up observation and vigilance with certain monitoring systems will reduce the number of mishaps.6 For an anesthesiologist to ignore the need for such a backup system is folly Hypoxic accidents, for example, still occur, resulting in multimillion-dollar awards against hospitals and anesthesiologists as well as untold suffering for patients and their families.
It is of utmost importance, however, for all parties-the public, attorneys, insurance companies, the government, and physicians-to recognize that by no means are all adverse events during anesthesia a result of human error and therefore preventable. Rather, as Keats has stated, an unknown number are due to patient disease, idiosyncrasy, or other factors: anesthetists must not be unfairly blamed for these.9 Tinker and Roberts agree:
The idea that if a patient dies or sustains injury, error must have played a causal role has spawned much legal profit Juries are led to believe, and indeed often want to believe, that any untoward outcome must be rooted in physician fault. Although this misbelief is exploited by lawyers, it is rooted in long-standing self-deprecatory thinking within anesthesiology in particular, but also by medicine in general.
Moreover, the extensive attention that the anesthesia community pays to the review of anesthesia mishaps probably contributes to the increased number-of suits filed.
Lastly, it is appropriate to stress that as anesthetic techniques continue to improve, sicker and sicker patients are successfully undergoing increasingly difficult and complicated surgery, such as open heart procedures and organ transplants. Authorities here and abroad agree that anesthesia morbidity and mortality have declined significantly in recent years.
Major Impediments to Change
The following are some of the problems preventing further effective reduction of avoidable anesthesia mishaps.
In the United States-in contrast with Great Britain and Australia, for example-satisfactory reporting systems, mandatory or otherwise, do not exist. Current data, meager at best, comes largely from analysis of insurance claim files and critical incident studies.7,8,11
The difference in practice patterns among anesthesiologists and the variability of their opinions concerning methods to reduce morbidity and mortality make it difficult to reach a consensus regarding standards of practice in anesthesia. Many authorities advocate the installation of sophisticated delivery and monitoring systems (as the airline industry has done), whereas others strongly oppose this approach and are convinced that safe anesthesia is a result of constant vigilance using relatively simple techniques.4,6,12,13
Cost containment is a major factor, if not the major factor, preventing adequate development, purchase, and maintenance of state-of-the-art equipment and systems.4
For an individual anesthetist or hospital, anesthetic mishaps are rare events. This fact greatly reduces continued interest in the problem among hospital medical staff and administrators. However, no matter how unusual an anesthetic death or severe injury may be, an individual occurrence is absolutely significant for that patient and family.
Few institutions-hospitals, anesthesia departments, professional associations, insurance companies, government agencies-are paying enough attention to these problems. However, a number of the issues discussed in this article are currently being addressed by a Task Force on Surgery and Anesthesia whose members represent the Joint Commission on Accreditation of Hospitals, the American Society of Anesthesiologists (ASA), the American College of Surgeons, the Association of Operating Room Nurses, the American Dental Association, the American Academy of Pediatrics, and other organizations.
Potential Solutions
In examining some of the potential solutions, it should be emphasized that funds needed for their implementation are not great when compared to the costs of medical liability and expenditures for long term care of the injured.
Compulsory reporting
It is important to establish mandatory systems for reporting anesthesia-related morbidity and mortality at local (hospital), regional, and national levels to provide data on the types and frequency of problems. This would allow anesthesia departments to take corrective action in risk management meetings examining mishaps, would encourage manufacturers to develop safer apparatus, would suggest that insurance companies reduce premiums for safe anesthetists, and would possibly allow anesthetists to choose anesthesia techniques on a statistically safer basis. The Joint Commission does require anesthesia departments to monitor and evaluate the quality and appropriateness of anesthesia patient care and to resolve identified problems.14
Standards
Most authorities believe that there should be a greater local and national emphasis on writing standards for the practice of anesthesia. Such standards might call for the development and use of safer anesthesia machines with emphasis on human4actor design issues and safety-performance criteria as well as a better integrated monitoring systems(6).
Actually, at least three sets of standards have been promulgated in recent months. These include those of the physician-owned Mutual Insurance Company of Arizona, the Harvard Medical School Department of Anesthesia, and Hospital Shared Services, Inc, the risk management division of MultiMedical Insurance Company. Descriptions of the first two are available in an ASA pamphlet. “Examples of Approaches to Risk Management.” In addition, proposed standards drafted by the joint Commission’s Task Force on Surgery and Anesthesia are being circulated for review. The ASA has also established an Ad Hoc Committee on Standards of Care to consider the problem.
Monitoring
Many authorities believe that routine monitoring should include electrocardiograms, oxygen analyzers, disconnect and high-pressure airway alarms, automatic blood pressure recording devices (noninvasive whenever satisfactory), temperature thermistors, pulse oximeters, and capnometers. They recognize that some of these devices cannot be used with every anesthetic and that monitoring does not replace but rather augments vigilance on the part of the anesthetist. Any anesthetist can make a mistake that may be detected by an alarm before harm occurs.
Evidence now demonstrates that, for high-risk patients, invasive hemodynamic and respiratory monitoring as well as other specific techniques, accompanied by aggressive treatment, significantly decrease morbidity and mortality after anesthesia and surgery.(15,16) It is just as important, however, to establish standards for routine, noninvasive monitoring to promote safety in low-risk patients.
The question of what to monitor routinely during every administration of anesthesia is most difficult to answer because of differences in opinion among well-qualified anesthesiologists. When assessing the relative cost-effectiveness of commonly available monitoring devices (fail-safe technology) to detect events such as esophageal intubation, disconnection in the breathing circuit, and administration of a hypoxic gas mixture, Duberman and Bendixen found that the recommended equipment costs only about $7 per patient over the useful life of the instruments.6 Clearly, compared with the staggering costs of medical liability insurance, that figure is minuscule. Of course, a fail-safe monitoring system may present its own problems. For example, when alarm systems are not integrated, warnings tend to be ignored; and when multiple units are attached, leaks may develop in the breathing system. Fully integrated systems, now becoming available, should provide greater safety.
It is appropriate here to mention specifically the recently marketed pulse oximeter, the capnometer, and devices for automated noninvasive measurement of blood pressure. Anesthetists are surprised by how often the oximeter registers desaturation during endotracheal intubation, after accumulation of bronchial secretions, with nitrous oxide washout, or upon arrival of the patient in the recovery room. Examples abound of large liability awards granted because of hypoxic events that could have been prevented by use of this device.
Oximetry should, however, be considered only as a supplement to techniques that more rapidly warn of inadequate inspired oxygen concentration, breathing circuit disconnects, and esophageal intubation. Capnography, for example, will indicate esophageal intubation within one breath, whereas it takes the pulse oximeter up to several minutes if alveolar oxygen pressure was previously high. Since hypertension and hypotension in patients with coronary artery disease are known to be associated with perioperative myocardial infarction, the more frequent determination of blood pressure that occurs with the use of automated blood-pressure-measuring apparatus after induction of anesthesia and during unstable periods may well decrease the incidence of infarction.
Certainly, vigilance is most important in anesthesiology; monitoring should never be more than a backup for the clinical acumen of the anesthesiologist. But we must recognize the major contributions that such monitoring devices can provide in reducing overall morbidity and mortality rates.
Multidisciplinary Risk Modification
Comparatively little is being spent by either government or professional organizations to make anesthesia safer. (R)educed anesthesia mortality rates can be cost-effectively achieved, but only through the organized, coordinated efforts of hospital professionals, (ICAH), malpractice insurers, and the anesthesia community itself groups that rarely work cooperatively on patient care issues.(4)
Problems of anesthesia mishaps should, therefore, be vigorously attacked at every level.
Hospitals and departments of anesthesia should do the following:
*Include anesthesia equipment in risk management programs. This will require periodic testing, documentation, and reporting of problems; prompt repair by professional technicians under the supervision of biomedical engineers; mandatory use of preoperative checklists by operators; and well-designed, inservice training programs.
* Replace obsolete anesthesia machines and monitoring equipment. Some authorities estimate ten years to be the safe, useful life of an anesthesia machine. Certainly many old machines do not meet modern safety standards. ECRI states, “Anesthetists must often tolerate a wide variety of equipment, as well as deficiencies such as dangerously arranged controls and gauges conditions that would not be tolerated. . in commercial aviation, regardless of the costs of correcting the deficiencies.4
* Establish effective morbidity and mortality conferences using data collected from listing all critical incidents. In addition, devote meetings periodically to overall patient safety issues.
* Emphasize in educational endeavors those engineering and functional aspects of anesthesia equipment that will result in better recognition of impending mishaps and the development of methods to prevent them.
Develop standards of practice.
Develop specific protocols, including guidelines for the exchange of anesthesia personnel and for anesthesia machine inspection procedures.
* Maintain rigid criteria for granting privileges to practice anesthesia.
* Employ enough technical anesthesia personnel to maintain equipment and provide assistance to the anesthetist when needed.
The ASA, the American Hospital Association, the American Medical Association, the Council of Medical Specialty Societies, and other concerned organizations should intensify their efforts to promote patient safety through strong educational programs for their members as well as continued evaluation of the effect of medical mishaps on ever-increasing medical liability premiums.
Medical liability insurance companies should lower premiums for anesthesiologists and hospitals that have effective anesthesia risk management programs, use superior monitoring systems, and follow proper standards of practice. In addition, insurance companies with adequate statistics should be able to recognize which departments or individual anesthetists sustain repeated mishaps and consider the addition of premium surcharges.
The Joint Commission should continue its current interest in anesthesia safety and its collaborative effort with the ASA and the other members of the Task Force on Surgery and Anesthesia to develop written standards for the practice of anesthesiology. Unfortunately, only by regulation will some anesthesia departments develop adequate standards of safety.
Government and other organizations should hasten the further development of safety and performance standards for anesthesia devices. This includes the establishment of standards for labeling ampules, vials, and syringes as proposed by the American Society for Testing and Materials Subcommittee D10.34.
Local and state governments should recognize anesthesia morbidity and mortality as a public health problem and establish statistical reporting systems. The Food and Drug Administration has issued a regulation requiring manufacturers and importers to report device-related deaths and serious injuries. It should enlarge its reporting system so that practitioners can anonymously record adverse anesthesia events.
Conclusion
The anesthesia profession shares many similarities with the airline industry, which has simply taken better advantage of risk management techniques and other technological advances. These include preflight checklists and other protocols, training in management of emergencies’ with “in-f light simulators,” and recording and tabulating critical incidents by means of the “black box” in-f light recorder. Perhaps this is why in the United States the average annual death rate from 1975 to 1980 for majorcommercial jetflights was only 177 per year.
In a January 28,1985, article about the disaster at the chemical plant in Bhopal, India, the New York Times noted that the accident resulted from design flaws, maintenance failures, training deficiencies, and operating errors. This is a good analogy of anesthesia mishaps, which also occur as a result of complex events with multiple possible causes. Only by improving our systems of training, monitoring, maintenance, and reporting, as well as by increasing our attention to human error, can we make anesthesia completely safe. This will require the vigorous efforts and close cooperation of all involved parties.
References
1. Snow J: On the fatal cases of the inhalation of chloroform. Edinburgh Med Surg 1 72:75,1849.
2. Pierce EC Jr: historical perspectives. In Pierce EC Jr, Cooper JB (eds): Analysis of Anesthetic Mishaps. International Anesthesiology Clinics. Boston: Little, Brown, 1984, pp 1-16.
3. Keenan RL. Boyan P: Cardiac arrest due to anesthesia: A study of incidence and causes. ]AMA 253:2373-2377, Apr 26,1985.
4. ECRI: Deaths during general anesthesia: Technology-related, due to human error, or unavoidable? Technology for Anesthesia 5:1-10, Mar 1985.
5. Pierce EC Jr: Reducing preventable anesthesia mishaps: A need for greater risk management initiatives. Risk Management Foundation Forum 6:6-8, Mar-Apr 1985.
6. Duberman SM, Bendixen HH: Concepts of fail-safe anesthetic practice. In Pierce EC Jr, Cooper JB (eds): Analysis of Anesthetic Mishaps. International Anesthesiology Clinics. Boston: Little, Brown, 1984, pp 149165.
7. Davis DA: An analysis of anesthetic mishaps from medical liability claims. In Pierce EC Jr, Cooper JB (eds): Analysis of Anesthetic Mishaps. International Anesthesiology Clinics. Boston: Little, Brown, 1984, pp 31-42.
8. Solazzi RW, Ward Rj: The spectrum of medical liability cases. In Pierce EC Jr, Cooper IB (eds): Analysis of Anesthetic Mishaps. International Anesthesiology Clinics. Boston: Little, Brown, 1984, pp 43-59.
9. Keats AS: Role of anesthesia in surgical mortality. In Orkin FK, Cooperman LH (eds): Complications in Anesthesiology. Philadelphia: Lippincott, 1983, pp 3-13.
10. Tinker IH, Roberts SL: Anesthesia risk. In Miller RD (ed): Anesthesia, 2d ed, vol 1. New York: Churchill Livingstone, 1986, pp 359-380.
11. Cooper IB, et al: Critical incidents associated with intraoperative changes of anesthesia personnel. Anesthesiology 65:456-461, Jun 1982.
12. Philip J H, Raemer DB: Selecting the optimal anesthesia monitoring array. Medical Instrumentation 19:122-126, May-Jun 1985.
13. Pierce EC Jr: Editorial: Standards for monitoring. Intelligence Reports in Anesthesia 3:3, Sep-Oct 1985.
14. Joint Commission on Accreditation of Hospitals (JCAH): Accreditation Manual for Hospitals. Chicago: JCAH, 1986, p 8.
15. RaoRLK,JacobsKJ,EI-EtrAA:Reinfarction following anesthesia in patients with myocardial infarctions. Anesthesiology 59:499-505, Dec 1985.
16. Slogoff S, Keats AS: Does perioperative myocardial ischemia lead to postoperative myocardial infarction? Anesthesiology 62:107-114, Feb 1985.
Reprinted with permission from Risk Management and Quality Assurance, a 152-page compendium of 26 related articles and a special publication from the Quality Review Bulletin, which is published by the
Joint Commission on Accreditation of Hospitals, 875 North Michigan Ave., Chicago, IL 60611, (312) 642-6061.